INDEX

Explanations

mentions of actors, especially in the context of their work

oai_token-act-pair · gpt-3.5-turbo

mentions of "actor" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

£ı

-0.81

outine

-0.70

plings

-0.67

wart

-0.67

vironment

-0.65

ruciating

-0.64

elsius

-0.63

xit

-0.62

yip

-0.62

asonable

-0.61

POSITIVE LOGITS

rities

1.03

 portraying

0.90

 actor

0.85

acters

0.83

writers

0.83

 Natalie

0.77

 actress

0.76

plays

0.73

 actresses

0.72

ãĤ´

0.70

Activations Density 0.028%

mentions of actors, especially in the context of their work

mentions of "actor" in various contexts

No Comments

No Known Activations